linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Glauber Costa <glommer@parallels.com>
To: linux-mm@kvack.org
Cc: linux-fsdevel@vger.kernel.org,
	containers@lists.linux-foundation.org,
	Michal Hocko <mhocko@suse.cz>,
	Johannes Weiner <hannes@cmpxchg.org>,
	kamezawa.hiroyu@jp.fujitsu.com,
	Andrew Morton <akpm@linux-foundation.org>,
	Dave Shrinnker <david@fromorbit.com>,
	Greg Thelen <gthelen@google.com>,
	hughd@google.com, yinghan@google.com,
	Dave Chinner <dchinner@redhat.com>
Subject: [PATCH v2 11/28] list_lru: per-node list infrastructure
Date: Fri, 29 Mar 2013 13:13:53 +0400	[thread overview]
Message-ID: <1364548450-28254-12-git-send-email-glommer@parallels.com> (raw)
In-Reply-To: <1364548450-28254-1-git-send-email-glommer@parallels.com>

From: Dave Chinner <dchinner@redhat.com>

Now that we have an LRU list API, we can start to enhance the
implementation.  This splits the single LRU list into per-node lists
and locks to enhance scalability. Items are placed on lists
according to the node the memory belongs to. To make scanning the
lists efficient, also track whether the per-node lists have entries
in them in a active nodemask.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
---
 include/linux/list_lru.h |  14 +++--
 lib/list_lru.c           | 160 +++++++++++++++++++++++++++++++++++------------
 2 files changed, 129 insertions(+), 45 deletions(-)

diff --git a/include/linux/list_lru.h b/include/linux/list_lru.h
index 3423949..b0e3ba25 100644
--- a/include/linux/list_lru.h
+++ b/include/linux/list_lru.h
@@ -8,21 +8,23 @@
 #define _LRU_LIST_H 0
 
 #include <linux/list.h>
+#include <linux/nodemask.h>
 
-struct list_lru {
+struct list_lru_node {
 	spinlock_t		lock;
 	struct list_head	list;
 	long			nr_items;
+} ____cacheline_aligned_in_smp;
+
+struct list_lru {
+	struct list_lru_node	node[MAX_NUMNODES];
+	nodemask_t		active_nodes;
 };
 
 int list_lru_init(struct list_lru *lru);
 int list_lru_add(struct list_lru *lru, struct list_head *item);
 int list_lru_del(struct list_lru *lru, struct list_head *item);
-
-static inline long list_lru_count(struct list_lru *lru)
-{
-	return lru->nr_items;
-}
+long list_lru_count(struct list_lru *lru);
 
 typedef int (*list_lru_walk_cb)(struct list_head *item, spinlock_t *lock,
 				void *cb_arg);
diff --git a/lib/list_lru.c b/lib/list_lru.c
index 475d0e9..881e342 100644
--- a/lib/list_lru.c
+++ b/lib/list_lru.c
@@ -6,6 +6,7 @@
  */
 #include <linux/kernel.h>
 #include <linux/module.h>
+#include <linux/mm.h>
 #include <linux/list_lru.h>
 
 int
@@ -13,14 +14,19 @@ list_lru_add(
 	struct list_lru	*lru,
 	struct list_head *item)
 {
-	spin_lock(&lru->lock);
+	int nid = page_to_nid(virt_to_page(item));
+	struct list_lru_node *nlru = &lru->node[nid];
+
+	spin_lock(&nlru->lock);
+	BUG_ON(nlru->nr_items < 0);
 	if (list_empty(item)) {
-		list_add_tail(item, &lru->list);
-		lru->nr_items++;
-		spin_unlock(&lru->lock);
+		list_add_tail(item, &nlru->list);
+		if (nlru->nr_items++ == 0)
+			node_set(nid, lru->active_nodes);
+		spin_unlock(&nlru->lock);
 		return 1;
 	}
-	spin_unlock(&lru->lock);
+	spin_unlock(&nlru->lock);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(list_lru_add);
@@ -30,43 +36,72 @@ list_lru_del(
 	struct list_lru	*lru,
 	struct list_head *item)
 {
-	spin_lock(&lru->lock);
+	int nid = page_to_nid(virt_to_page(item));
+	struct list_lru_node *nlru = &lru->node[nid];
+
+	spin_lock(&nlru->lock);
 	if (!list_empty(item)) {
 		list_del_init(item);
-		lru->nr_items--;
-		spin_unlock(&lru->lock);
+		if (--nlru->nr_items == 0)
+			node_clear(nid, lru->active_nodes);
+		BUG_ON(nlru->nr_items < 0);
+		spin_unlock(&nlru->lock);
 		return 1;
 	}
-	spin_unlock(&lru->lock);
+	spin_unlock(&nlru->lock);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(list_lru_del);
 
 long
-list_lru_walk(
-	struct list_lru *lru,
-	list_lru_walk_cb isolate,
-	void		*cb_arg,
-	long		nr_to_walk)
+list_lru_count(
+	struct list_lru *lru)
 {
+	long count = 0;
+	int nid;
+
+	for_each_node_mask(nid, lru->active_nodes) {
+		struct list_lru_node *nlru = &lru->node[nid];
+
+		spin_lock(&nlru->lock);
+		BUG_ON(nlru->nr_items < 0);
+		count += nlru->nr_items;
+		spin_unlock(&nlru->lock);
+	}
+
+	return count;
+}
+EXPORT_SYMBOL_GPL(list_lru_count);
+
+static long
+list_lru_walk_node(
+	struct list_lru		*lru,
+	int			nid,
+	list_lru_walk_cb	isolate,
+	void			*cb_arg,
+	long			*nr_to_walk)
+{
+	struct list_lru_node	*nlru = &lru->node[nid];
 	struct list_head *item, *n;
-	long removed = 0;
+	long isolated = 0;
 restart:
-	spin_lock(&lru->lock);
-	list_for_each_safe(item, n, &lru->list) {
+	spin_lock(&nlru->lock);
+	list_for_each_safe(item, n, &nlru->list) {
 		int ret;
 
-		if (nr_to_walk-- < 0)
+		if ((*nr_to_walk)-- < 0)
 			break;
 
-		ret = isolate(item, &lru->lock, cb_arg);
+		ret = isolate(item, &nlru->lock, cb_arg);
 		switch (ret) {
 		case 0:	/* item removed from list */
-			lru->nr_items--;
-			removed++;
+			if (--nlru->nr_items == 0)
+				node_clear(nid, lru->active_nodes);
+			BUG_ON(nlru->nr_items < 0);
+			isolated++;
 			break;
 		case 1: /* item referenced, give another pass */
-			list_move_tail(item, &lru->list);
+			list_move_tail(item, &nlru->list);
 			break;
 		case 2: /* item cannot be locked, skip */
 			break;
@@ -76,42 +111,89 @@ restart:
 			BUG();
 		}
 	}
-	spin_unlock(&lru->lock);
-	return removed;
+	spin_unlock(&nlru->lock);
+	return isolated;
+}
+
+long
+list_lru_walk(
+	struct list_lru	*lru,
+	list_lru_walk_cb isolate,
+	void		*cb_arg,
+	long		nr_to_walk)
+{
+	long isolated = 0;
+	int nid;
+
+	for_each_node_mask(nid, lru->active_nodes) {
+		isolated += list_lru_walk_node(lru, nid, isolate,
+					       cb_arg, &nr_to_walk);
+		if (nr_to_walk <= 0)
+			break;
+	}
+	return isolated;
 }
 EXPORT_SYMBOL_GPL(list_lru_walk);
 
 long
-list_lru_dispose_all(
-	struct list_lru *lru,
-	list_lru_dispose_cb dispose)
+list_lru_dispose_all_node(
+	struct list_lru		*lru,
+	int			nid,
+	list_lru_dispose_cb	dispose)
 {
-	long disposed = 0;
+	struct list_lru_node	*nlru = &lru->node[nid];
 	LIST_HEAD(dispose_list);
+	long disposed = 0;
 
-	spin_lock(&lru->lock);
-	while (!list_empty(&lru->list)) {
-		list_splice_init(&lru->list, &dispose_list);
-		disposed += lru->nr_items;
-		lru->nr_items = 0;
-		spin_unlock(&lru->lock);
+	spin_lock(&nlru->lock);
+	while (!list_empty(&nlru->list)) {
+		list_splice_init(&nlru->list, &dispose_list);
+		disposed += nlru->nr_items;
+		nlru->nr_items = 0;
+		node_clear(nid, lru->active_nodes);
+		spin_unlock(&nlru->lock);
 
 		dispose(&dispose_list);
 
-		spin_lock(&lru->lock);
+		spin_lock(&nlru->lock);
 	}
-	spin_unlock(&lru->lock);
+	spin_unlock(&nlru->lock);
 	return disposed;
 }
 
+long
+list_lru_dispose_all(
+	struct list_lru		*lru,
+	list_lru_dispose_cb	dispose)
+{
+	long disposed;
+	long total = 0;
+	int nid;
+
+	do {
+		disposed = 0;
+		for_each_node_mask(nid, lru->active_nodes) {
+			disposed += list_lru_dispose_all_node(lru, nid,
+							      dispose);
+		}
+		total += disposed;
+	} while (disposed != 0);
+
+	return total;
+}
+
 int
 list_lru_init(
 	struct list_lru	*lru)
 {
-	spin_lock_init(&lru->lock);
-	INIT_LIST_HEAD(&lru->list);
-	lru->nr_items = 0;
+	int i;
 
+	nodes_clear(lru->active_nodes);
+	for (i = 0; i < MAX_NUMNODES; i++) {
+		spin_lock_init(&lru->node[i].lock);
+		INIT_LIST_HEAD(&lru->node[i].list);
+		lru->node[i].nr_items = 0;
+	}
 	return 0;
 }
 EXPORT_SYMBOL_GPL(list_lru_init);
-- 
1.8.1.4

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  parent reply	other threads:[~2013-03-29  9:14 UTC|newest]

Thread overview: 97+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-03-29  9:13 [PATCH v2 00/28] memcg-aware slab shrinking Glauber Costa
2013-03-29  9:13 ` [PATCH v2 01/28] super: fix calculation of shrinkable objects for small numbers Glauber Costa
2013-04-01  7:16   ` Kamezawa Hiroyuki
2013-03-29  9:13 ` [PATCH v2 02/28] vmscan: take at least one pass with shrinkers Glauber Costa
2013-04-01  7:26   ` Kamezawa Hiroyuki
2013-04-01  8:10     ` Glauber Costa
2013-04-10  5:09       ` Ric Mason
2013-04-10  7:32         ` Glauber Costa
2013-04-10  9:19         ` Dave Chinner
2013-04-08  8:42   ` Joonsoo Kim
2013-04-08  8:47     ` Glauber Costa
2013-04-08  9:01       ` Joonsoo Kim
2013-04-08  9:05         ` Glauber Costa
2013-04-09  0:55           ` Joonsoo Kim
2013-04-09  1:29             ` Dave Chinner
2013-04-09  2:05               ` Joonsoo Kim
2013-04-09  7:43                 ` Glauber Costa
2013-04-09  9:08                   ` Joonsoo Kim
2013-04-09 12:30                 ` Dave Chinner
2013-04-10  2:51                   ` Joonsoo Kim
2013-04-10  7:30                     ` Glauber Costa
2013-04-10  8:19                       ` Joonsoo Kim
2013-04-10  8:46                     ` Wanpeng Li
2013-04-10  8:46                     ` Wanpeng Li
2013-04-10 10:07                       ` Dave Chinner
2013-04-10 14:03                         ` JoonSoo Kim
2013-04-11  0:41                           ` Dave Chinner
2013-04-11  7:27                             ` Wanpeng Li
2013-04-11  7:27                             ` Wanpeng Li
2013-04-11  7:27                             ` Wanpeng Li
2013-04-11  9:25                               ` Dave Chinner
     [not found]                     ` <20130410025115.GA5872-Hm3cg6mZ9cc@public.gmane.org>
2013-04-10  8:46                       ` Wanpeng Li
2013-03-29  9:13 ` [PATCH v2 03/28] dcache: convert dentry_stat.nr_unused to per-cpu counters Glauber Costa
2013-04-05  1:09   ` Greg Thelen
2013-04-05  1:15     ` Dave Chinner
2013-04-08  9:14       ` Glauber Costa
2013-04-08 13:18         ` Glauber Costa
2013-04-08 23:26         ` Dave Chinner
2013-04-09  8:02           ` Glauber Costa
2013-04-09 12:47             ` Dave Chinner
2013-03-29  9:13 ` [PATCH v2 04/28] dentry: move to per-sb LRU locks Glauber Costa
2013-03-29  9:13 ` [PATCH v2 05/28] dcache: remove dentries from LRU before putting on dispose list Glauber Costa
2013-04-03  6:51   ` Sha Zhengju
2013-04-03  8:55     ` Glauber Costa
2013-04-04  6:19     ` Dave Chinner
2013-04-04  6:56       ` Glauber Costa
2013-03-29  9:13 ` [PATCH v2 06/28] mm: new shrinker API Glauber Costa
2013-04-05  1:09   ` Greg Thelen
2013-03-29  9:13 ` [PATCH v2 07/28] shrinker: convert superblock shrinkers to new API Glauber Costa
2013-03-29  9:13 ` [PATCH v2 08/28] list: add a new LRU list type Glauber Costa
2013-04-04 21:53   ` Greg Thelen
2013-04-05  1:20     ` Dave Chinner
2013-04-05  8:01       ` Glauber Costa
2013-04-06  0:04         ` Dave Chinner
2013-03-29  9:13 ` [PATCH v2 09/28] inode: convert inode lru list to generic lru list code Glauber Costa
2013-03-29  9:13 ` [PATCH v2 10/28] dcache: convert to use new lru list infrastructure Glauber Costa
2013-04-08 13:14   ` Glauber Costa
2013-04-08 23:28     ` Dave Chinner
2013-03-29  9:13 ` Glauber Costa [this message]
2013-03-29  9:13 ` [PATCH v2 12/28] shrinker: add node awareness Glauber Costa
2013-03-29  9:13 ` [PATCH v2 13/28] fs: convert inode and dentry shrinking to be node aware Glauber Costa
2013-03-29  9:13 ` [PATCH v2 14/28] xfs: convert buftarg LRU to generic code Glauber Costa
2013-03-29  9:13 ` [PATCH v2 15/28] xfs: convert dquot cache lru to list_lru Glauber Costa
2013-03-29  9:13 ` [PATCH v2 16/28] fs: convert fs shrinkers to new scan/count API Glauber Costa
2013-03-29  9:13 ` [PATCH v2 17/28] drivers: convert shrinkers to new count/scan API Glauber Costa
2013-03-29  9:14 ` [PATCH v2 18/28] shrinker: convert remaining shrinkers to " Glauber Costa
2013-03-29  9:14 ` [PATCH v2 19/28] hugepage: convert huge zero page shrinker to new shrinker API Glauber Costa
2013-03-29  9:14 ` [PATCH v2 20/28] shrinker: Kill old ->shrink API Glauber Costa
2013-03-29  9:14 ` [PATCH v2 21/28] vmscan: also shrink slab in memcg pressure Glauber Costa
2013-04-01  7:46   ` Kamezawa Hiroyuki
2013-04-01  8:51     ` Glauber Costa
2013-04-03 10:11   ` Sha Zhengju
2013-04-03 10:43     ` Glauber Costa
2013-04-04  9:35       ` Sha Zhengju
2013-04-05  8:25         ` Glauber Costa
2013-03-29  9:14 ` [PATCH v2 22/28] memcg,list_lru: duplicate LRUs upon kmemcg creation Glauber Costa
2013-04-01  8:05   ` Kamezawa Hiroyuki
2013-04-01  8:22     ` Glauber Costa
2013-03-29  9:14 ` [PATCH v2 23/28] lru: add an element to a memcg list Glauber Costa
2013-04-01  8:18   ` Kamezawa Hiroyuki
2013-04-01  8:29     ` Glauber Costa
2013-03-29  9:14 ` [PATCH v2 24/28] list_lru: also include memcg lists in counts and scans Glauber Costa
2013-03-29  9:14 ` [PATCH v2 25/28] list_lru: per-memcg walks Glauber Costa
2013-03-29  9:14 ` [PATCH v2 26/28] memcg: per-memcg kmem shrinking Glauber Costa
2013-04-01  8:31   ` Kamezawa Hiroyuki
2013-04-01  8:48     ` Glauber Costa
2013-04-01  9:01       ` Kamezawa Hiroyuki
2013-04-01  9:14         ` Glauber Costa
2013-04-01  9:35         ` Kamezawa Hiroyuki
2013-03-29  9:14 ` [PATCH v2 27/28] list_lru: reclaim proportionaly between memcgs and nodes Glauber Costa
2013-03-29  9:14 ` [PATCH v2 28/28] super: targeted memcg reclaim Glauber Costa
2013-04-01 12:38 ` [PATCH v2 00/28] memcg-aware slab shrinking Serge Hallyn
2013-04-01 12:45   ` Glauber Costa
2013-04-01 14:12     ` Serge Hallyn
2013-04-08  8:11       ` Glauber Costa
2013-04-02  4:58   ` Dave Chinner
2013-04-02  7:55     ` Glauber Costa

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1364548450-28254-12-git-send-email-glommer@parallels.com \
    --to=glommer@parallels.com \
    --cc=akpm@linux-foundation.org \
    --cc=containers@lists.linux-foundation.org \
    --cc=david@fromorbit.com \
    --cc=dchinner@redhat.com \
    --cc=gthelen@google.com \
    --cc=hannes@cmpxchg.org \
    --cc=hughd@google.com \
    --cc=kamezawa.hiroyu@jp.fujitsu.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@suse.cz \
    --cc=yinghan@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox